118 DEVELOPMENT OF COMPOUND 



hippocastanum), Fig. 5, the separation of tlie leaflets has been 

 carried down to the summit of the main petiole or support, 

 for each leaflet is fully formed and therefore distinct from the 

 leaflets on either side. The attention of the reader is called 

 to the correspondence which subsists between the coalescing 

 leaflets of the Castor-oil leaf and the free leaflets of the Horse 

 Chestnut leaf, in size and in direction. It is evident that the 

 general plan of structure is much the same in both, with this 

 difference, that in the Castor-oil the leaflets, being only par- 

 tially developed, still continue to coalesce with each other, 

 whilst in the Horse Chestnut each leaflet is fully formed, and 

 therefore separate and free. 



The passage of the digitate into the pinnate forms of com- 

 pound leaf, is simply effected by the formation of a naked 

 portion qf stem between each pair of leaflets. Compare, in 

 this respect, the leaf of the Horse Chestnut with that of the 

 White-Heart Hickory {Carya tomentosa), Fig. 6. There is 

 precisely the same correspondence subsisting between the 

 leaflets of this digitate and pinnate variety of compound leaf, 

 in size and in direction, as in the former case. We are justi- 

 fied, therefore in regarding the pinnate leaf as differing from 

 the leaf that is digitate, only in the formation of those little 

 intervals of midrib which separate its several pairs of leaflets ; 

 for if we suppose them to become rudimentary, then these 

 leaflets become crowded together at the top of a common 

 petiole, and the pijjnate at once passes into the digitate form, 

 from which it can no longer be distinguished. 



The transition of the pinnate .into the bipinnate form is a 

 fa6t which has been long known to the world. This is well 

 seen in the leaves of the common Locust-tree {Rohinia pseuda- 

 cacia), Fig. 7, where, through the increased development of 

 the costa, a, and its lateral fibres, b, the new costse, c, and a 

 second set of leaflets, d, is produced. This second set of 

 leaflets may in like manner produce another set, which- is the 

 case in the tripinnate variety of compound leaf; but here the 

 vegetative power reaches its limit, for it is seldom, that the 

 division goes beyond the third degree, although tripinnate 

 leaves with their leaflets pinnatified are not tmcommon. 



